You realise, I trust, that one of the reasons why people moved away from metal-backed currencies is that they were too volatile? A new large deposit is found, the currency inflates. A new high-demand use is found, the currency deflates. A more efficient material replacing it for a large industrial uses, the currency inflates.
There is no such thing as stealth in space. Anything coming from the moon to the Earth would either need a huge amount of thrust to go straight down (which would be easy to detect), or would cross into daylight and reflect a lot of heat even if it were black. And you can bet that anyone who had a launch capability on the moon would be under close scrutiny.
They'd also better have a first-rate missile defence shield, because in the two days it would take their projectile to reach Earth, their target would launch a lot of ballistic missiles (which take a maximum of about 40 minutes to arrive, anywhere in the world) and render their cities smouldering ruins.
And if they're relying on a large mass, not a warhead, then they won't have much delta-V for dodging, so they'll be easy to intercept. If they've painted it black, then it will be very vulnerable to heating by laser weapons and whatever colour it is it will be easy to fragment. The kind of mass that you'd need to survive reentry would either need to be huge or covered in (very fragile and vulnerable to attack) ceramic tiles to dissipate the heat. If you fragment it, the atmospheric friction will cause the entire thing to evaporate.
If attacking from space were easy, then Earth would have had a lot more extinction events in its past. The atmosphere alone does a pretty good job as a meteor shield.
You don't even need to cut the military that much. Cutting back 1/3 would cover the US debt interest payments.
For reference, cutting 1/3 of US military spending would move the US from second place (behind Saudi Arabia) to third (also behind Russia) in terms of defence spending as a percentage of GDP, and still leave them top in absolute terms (spending three times as much as the PRC, in second place) and top in per-capita spending.
It's not entirely clear, however, that reducing military spending would help the US economy. A lot of R&D is subsidised from the military budget, which helps drive US high tech exports, and you can bet that if you cur the military budget by a third then this would be the first place that congress would want reductions...
The phone might be, but all of the charging infrastructure won't be. People don't want to have to replace docks and so on with something new in 2-3 years. A TV that docks a mobile phone and allows it to function as a computer with a big screen may last 10 years, even if the phone is replaced after 18 months. That's likely to be a pretty common use case in a few years, so if you design a standard now that doesn't support it then it's a failure. And, no, USB3 is not fast enough for HD video.
There's a crappy standard, and Apple is ignoring it because they have something better. You want everyone to implement the standard, you need to start with a good standard and then say 'don't use your inferior connector, use the standard'.
I agree. There definitely should be a standard connector for this kind of thing, but it should be something that doesn't suck. The old Apple Dock connectors had a lot more functionality than the newer lightning ones, but the connector was bit too big. A ubiquitous connector needs:
A future-proof data signal (e.g. Thunderbolt, which can carry a signal fast enough that it won't be obsolete within a couple of years of release), that doesn't need to be supported by endpoints but can be detected and used if it is.
A widely-supported legacy signal (e.g. USB) so that it works everywhere
A lightweight mechanism for negotiating power demands and capabilities between supply and device.
A physically sturdy connector, with a reference design of a socket that will stand at least 1,000 insertions and ideally 10,000 in normal use.
A connector that either has an orientation so obvious that no one could possibly plug it in the wrong way, or one that works in either orientation.
Any patents that cover the design must be licensed royalty free, so third parties can interface with it cheaply and easily.
Neither microUSB nor Lightning meets these requirements. If Nokia wants to fix this, they should get together an industry group to design and agree to use such a connector. Don't complain at Apple, design a better connector than the Apple one, get everyone except Apple behind it, and market the hell out of it. Make every non-Apple phone have a big sticker on it saying that it supports the standard connector and list the features that make it better than the Apple one.
In that case, neither Ubuntu nor Fedora is Linux. There's about as much glibc code as there is Linux code in one of those, and the amount of X.org code dwarfs either. Even if you strip it back to a non-graphical core that can just boot, there's more GNU code than Linux. If you take a system like Debian and replace the kernel with a FreeBSD kernel, most users won't notice the difference.
Why does that matter to me as a user or integrator? It still means that I am locked in to whatever vendor they choose for their DRM. If that vendor chooses not to support my platform, or decides that I am a competitor in some other business so refuses to give me distribution rights to their EME plugin, then I'm stuck.
This is the entire point of the original question in TFA. Netflix gets the ability to (slightly) more easily move between vendors for DRM. What do users get? Nothing. There is no requirement that OMA plugins be interoperable and there is no guarantee of a second source. If Netflix decides to use MS PlayReady, but MS decides that they don't want to support my device because it competes with the Surface or the XBox, then I'm in exactly the same situation as I was with Silverlight.
Okay, where is the DRM'd video available today that I can play back on every device that I own that has sufficient processing power and display capabilities for video playback? Where is the DRM'd audio that I can play back on every device that I own with speakers and enough processing power to decode compressed audio?
Today, I can view any standards-compliant web site on any computer (desktop, laptop, tablet) that I own. I don't need the content providers to invest in my platform of choice, as long as enough people (or people with enough money) want to use it, they are free to support the standards independent of who is providing the content. If I want to create and sell some appliance to view the web, I can do so without being locked in to a single vendor for any of the required code. With DRM, none of this is true.
Pretty much every game I play, every video I watch, and every song I listen to is available on torrent web sites already, and yet I still pay for them. DRM does nothing to stop the availability of pirated versions, but it does impact me when it means that I can't play them. I have a FreeBSD machine connected to my projector and surround sound system at home. It plays BBC iPlayer and DVDs fine. It won't play Netflix. Who loses? Netflix, because they're trying to sell a service that I would happily pay money for (I already pay more than the cost of a Netflix subscription to a different company for DVD rentals), but can't use because they choose not to support the platform I'm using.
Who else loses? Consumers, because while we have a large number of competing providers of MP3 players and TVs, we have a very restricted set of providers of who can create Netflix streaming devices. They all have to either build their systems on Windows and license Silverlight form Microsoft or directly negotiate with Netflix.
I stopped buying games when the copy protection became too annoying. I started again when GOG started selling DRM-free games in a (very) convenient UI. They've made a load of money from me because they sell games at a price I want to pay in a form that I'm willing to pay for. I have about a dozen games that I've bought from them and not had time to play yet.
It's just consolidating the APIs so that these providers can create HTML5 web apps that run on more devices without modification
Consolidating the APIs isn't worth a thing when the APIs are just talking to some OS-specific (and possibly browser-specific) blob, which is what the W3C is actually proposing. Who cares if Netflix is using an open API, if instead of using MS Silverlight they're now using MS DRM Plugin?
We bother with them because they have a stable government. Unfortunately, the easiest way to have a stable government is to oppress most of the population. Democracies have a habit of changing their foreign policy every few years, which isn't nearly as good when you want to build a import or export market.
China is socialist and not capitalist? Have you been asleep for the last few decades? Or do you also believe that the Democratic Republic of North Korea is actually democratic, and that the United States of America is actually united?
Most executives would be happy to say 'we'll accept a slightly lower margin if we can make up for it in sales volume because more people will buy our product if we can guarantee a certain standard of living for people on our production lines'. At least, they would be happy to say it if it were true. Unless people are willing to boycott Sony and not buy a PS4 over this kind of thing, they have no incentive to stop.
I really don't understand why Google invests so much in trying to build profiles of people. It's almost enough to make you believe that the NSA funded or pressured them to do it, because it makes no business sense at all.
When Google started, they had simple text-based ads that were placed because they were relevant to the text on the page. They were unobtrusive and they were - at least somewhat - informative (because you can't do much else with plain text and still make people want to click on it). I clicked on a lot of Google Ads back in those days, because they were almost always about things that I was directly interested in at the time I saw the ad.
Now, Google Ads are based on things that Google thinks I'm interested in based on my past behaviour and this means that they are pretty much completely irrelevant to me. I don't block ads, but it's been a long time since I saw one that I even vaguely considered clicking on.
In the 1940s, the US also had university subsidy for anyone who fought in the Second World War. That meant that by the 1950s, the US had a far higher proportion of its population with degrees than anywhere else in the world. That made it an attractive place to start companies that needed a lot of educated employees, which made it an attractive place for educated people to emigrate to (see: brain drain). Now, university education is expensive, attendance is dropping, and immigration policies mean that the US is harder for educated to enter than a number of other places with a high standard of living.
According to a BBC article, the American Dream is upwards social mobility: that your children will be better educated and better off than you were. The point of the article was that downward social mobility is increasingly common in the US.
It's more subtle than that. In the US, anyone travelling on the government's dime[1] has a fairly strict per diem for accommodation, and a more generous (relative to real costs) one for meals and incidentals. If the hotel charges for WiFi, then this comes out of the M&I budget. This is also why they often charge for breakfast separately. It means that they'll get government customers, even if their room+WiFi+breakfast rate is higher than someone else's, because their room rate is below the per diem while someone who includes all of the other stuff isn't.
Oh, and a lot of US hotels are very willing to take the rate down to the State Department's per diem rate if you ask them to, because they know that they'll get a lot of business if they have a reputation for being friendly to government employees.
[1] This includes both direct government employees and a lot of people funded indirectly, for example people working for a university employed on a project funded with a government grant.
For a good overview, take a look at Jakob Olesen's talk from EuroLLVM. Be warned, the talk contains a lot of simplifications (and is completely wrong if you're targeting anything like a GPU or DSP), but it's a good overview. For very recent chips and scientific workloads, it's also interesting that it often uses a lot less power (and is faster) to recompute intermediate results than to store them, if they won't fit in cache (sometimes even if they won't fit in L1 cache).
I teach a postgraduate compilers course, so I definitely wouldn't downplay the importance of understanding how compilers work. The problem is that a lot of people know how compilers worked in the '70s and '80s and the techniques that made sense for optimising for a PDP-11 or VAX (or even i386) and these techniques can often hurt performance today.
Slashdot Mirror
Did you even bother to read my post, where I stated there was no standard that actually met the requirements and listed what those were?
You realise, I trust, that one of the reasons why people moved away from metal-backed currencies is that they were too volatile? A new large deposit is found, the currency inflates. A new high-demand use is found, the currency deflates. A more efficient material replacing it for a large industrial uses, the currency inflates.
They'd also better have a first-rate missile defence shield, because in the two days it would take their projectile to reach Earth, their target would launch a lot of ballistic missiles (which take a maximum of about 40 minutes to arrive, anywhere in the world) and render their cities smouldering ruins.
And if they're relying on a large mass, not a warhead, then they won't have much delta-V for dodging, so they'll be easy to intercept. If they've painted it black, then it will be very vulnerable to heating by laser weapons and whatever colour it is it will be easy to fragment. The kind of mass that you'd need to survive reentry would either need to be huge or covered in (very fragile and vulnerable to attack) ceramic tiles to dissipate the heat. If you fragment it, the atmospheric friction will cause the entire thing to evaporate.
If attacking from space were easy, then Earth would have had a lot more extinction events in its past. The atmosphere alone does a pretty good job as a meteor shield.
You don't even need to cut the military that much. Cutting back 1/3 would cover the US debt interest payments.
For reference, cutting 1/3 of US military spending would move the US from second place (behind Saudi Arabia) to third (also behind Russia) in terms of defence spending as a percentage of GDP, and still leave them top in absolute terms (spending three times as much as the PRC, in second place) and top in per-capita spending.
It's not entirely clear, however, that reducing military spending would help the US economy. A lot of R&D is subsidised from the military budget, which helps drive US high tech exports, and you can bet that if you cur the military budget by a third then this would be the first place that congress would want reductions...
Yes, because it's so hard to buy a collocated server in China. You've not actually read any of the recent revelations about NSA practices, have you?
The phone might be, but all of the charging infrastructure won't be. People don't want to have to replace docks and so on with something new in 2-3 years. A TV that docks a mobile phone and allows it to function as a computer with a big screen may last 10 years, even if the phone is replaced after 18 months. That's likely to be a pretty common use case in a few years, so if you design a standard now that doesn't support it then it's a failure. And, no, USB3 is not fast enough for HD video.
There's a crappy standard, and Apple is ignoring it because they have something better. You want everyone to implement the standard, you need to start with a good standard and then say 'don't use your inferior connector, use the standard'.
Neither microUSB nor Lightning meets these requirements. If Nokia wants to fix this, they should get together an industry group to design and agree to use such a connector. Don't complain at Apple, design a better connector than the Apple one, get everyone except Apple behind it, and market the hell out of it. Make every non-Apple phone have a big sticker on it saying that it supports the standard connector and list the features that make it better than the Apple one.
Minrcraft is almost pure Java... So this should be easy, peasy, right?
I think people stopped believing that about a decade ago.
In that case, neither Ubuntu nor Fedora is Linux. There's about as much glibc code as there is Linux code in one of those, and the amount of X.org code dwarfs either. Even if you strip it back to a non-graphical core that can just boot, there's more GNU code than Linux. If you take a system like Debian and replace the kernel with a FreeBSD kernel, most users won't notice the difference.
Why does that matter to me as a user or integrator? It still means that I am locked in to whatever vendor they choose for their DRM. If that vendor chooses not to support my platform, or decides that I am a competitor in some other business so refuses to give me distribution rights to their EME plugin, then I'm stuck.
This is the entire point of the original question in TFA. Netflix gets the ability to (slightly) more easily move between vendors for DRM. What do users get? Nothing. There is no requirement that OMA plugins be interoperable and there is no guarantee of a second source. If Netflix decides to use MS PlayReady, but MS decides that they don't want to support my device because it competes with the Surface or the XBox, then I'm in exactly the same situation as I was with Silverlight.
Okay, where is the DRM'd video available today that I can play back on every device that I own that has sufficient processing power and display capabilities for video playback? Where is the DRM'd audio that I can play back on every device that I own with speakers and enough processing power to decode compressed audio?
Today, I can view any standards-compliant web site on any computer (desktop, laptop, tablet) that I own. I don't need the content providers to invest in my platform of choice, as long as enough people (or people with enough money) want to use it, they are free to support the standards independent of who is providing the content. If I want to create and sell some appliance to view the web, I can do so without being locked in to a single vendor for any of the required code. With DRM, none of this is true.
Pretty much every game I play, every video I watch, and every song I listen to is available on torrent web sites already, and yet I still pay for them. DRM does nothing to stop the availability of pirated versions, but it does impact me when it means that I can't play them. I have a FreeBSD machine connected to my projector and surround sound system at home. It plays BBC iPlayer and DVDs fine. It won't play Netflix. Who loses? Netflix, because they're trying to sell a service that I would happily pay money for (I already pay more than the cost of a Netflix subscription to a different company for DVD rentals), but can't use because they choose not to support the platform I'm using.
Who else loses? Consumers, because while we have a large number of competing providers of MP3 players and TVs, we have a very restricted set of providers of who can create Netflix streaming devices. They all have to either build their systems on Windows and license Silverlight form Microsoft or directly negotiate with Netflix.
I stopped buying games when the copy protection became too annoying. I started again when GOG started selling DRM-free games in a (very) convenient UI. They've made a load of money from me because they sell games at a price I want to pay in a form that I'm willing to pay for. I have about a dozen games that I've bought from them and not had time to play yet.
It's just consolidating the APIs so that these providers can create HTML5 web apps that run on more devices without modification
Consolidating the APIs isn't worth a thing when the APIs are just talking to some OS-specific (and possibly browser-specific) blob, which is what the W3C is actually proposing. Who cares if Netflix is using an open API, if instead of using MS Silverlight they're now using MS DRM Plugin?
We bother with them because they have a stable government. Unfortunately, the easiest way to have a stable government is to oppress most of the population. Democracies have a habit of changing their foreign policy every few years, which isn't nearly as good when you want to build a import or export market.
China is socialist and not capitalist? Have you been asleep for the last few decades? Or do you also believe that the Democratic Republic of North Korea is actually democratic, and that the United States of America is actually united?
Most executives would be happy to say 'we'll accept a slightly lower margin if we can make up for it in sales volume because more people will buy our product if we can guarantee a certain standard of living for people on our production lines'. At least, they would be happy to say it if it were true. Unless people are willing to boycott Sony and not buy a PS4 over this kind of thing, they have no incentive to stop.
According to my (1897) edition of Websters, ass has two meanings: donkey and stupid person. Maybe he meant you slap the donkey on her stupid person?
I really don't understand why Google invests so much in trying to build profiles of people. It's almost enough to make you believe that the NSA funded or pressured them to do it, because it makes no business sense at all.
When Google started, they had simple text-based ads that were placed because they were relevant to the text on the page. They were unobtrusive and they were - at least somewhat - informative (because you can't do much else with plain text and still make people want to click on it). I clicked on a lot of Google Ads back in those days, because they were almost always about things that I was directly interested in at the time I saw the ad.
Now, Google Ads are based on things that Google thinks I'm interested in based on my past behaviour and this means that they are pretty much completely irrelevant to me. I don't block ads, but it's been a long time since I saw one that I even vaguely considered clicking on.
In the 1940s, the US also had university subsidy for anyone who fought in the Second World War. That meant that by the 1950s, the US had a far higher proportion of its population with degrees than anywhere else in the world. That made it an attractive place to start companies that needed a lot of educated employees, which made it an attractive place for educated people to emigrate to (see: brain drain). Now, university education is expensive, attendance is dropping, and immigration policies mean that the US is harder for educated to enter than a number of other places with a high standard of living.
According to a BBC article, the American Dream is upwards social mobility: that your children will be better educated and better off than you were. The point of the article was that downward social mobility is increasingly common in the US.
It's more subtle than that. In the US, anyone travelling on the government's dime[1] has a fairly strict per diem for accommodation, and a more generous (relative to real costs) one for meals and incidentals. If the hotel charges for WiFi, then this comes out of the M&I budget. This is also why they often charge for breakfast separately. It means that they'll get government customers, even if their room+WiFi+breakfast rate is higher than someone else's, because their room rate is below the per diem while someone who includes all of the other stuff isn't.
Oh, and a lot of US hotels are very willing to take the rate down to the State Department's per diem rate if you ask them to, because they know that they'll get a lot of business if they have a reputation for being friendly to government employees.
[1] This includes both direct government employees and a lot of people funded indirectly, for example people working for a university employed on a project funded with a government grant.
For a good overview, take a look at Jakob Olesen's talk from EuroLLVM. Be warned, the talk contains a lot of simplifications (and is completely wrong if you're targeting anything like a GPU or DSP), but it's a good overview. For very recent chips and scientific workloads, it's also interesting that it often uses a lot less power (and is faster) to recompute intermediate results than to store them, if they won't fit in cache (sometimes even if they won't fit in L1 cache).
I teach a postgraduate compilers course, so I definitely wouldn't downplay the importance of understanding how compilers work. The problem is that a lot of people know how compilers worked in the '70s and '80s and the techniques that made sense for optimising for a PDP-11 or VAX (or even i386) and these techniques can often hurt performance today.